This thesis analyzes and explores solutions in which parallel and distributed high performance systems can improve the urgent needs of environmental community in managing unprecedented amount of environmental data to provide meaningful information regarding the changes that are threatening our environment and our society at large. We propose a general methodology and framework for easily porting and executing environmental applications simultaneously on different parallel and distributed infrastructures: cluster, Grid and Cloud. Such a Hybrid Computing Environment introduces complex challenges to tackle with, especially in supporting the interoperability and the coexistence of the underlying distributed, heterogeneous computing infrastructures. The interoperability between Environmental Sciences and such a Hybrid Computing Environment is also an important goal to achieve. The proposed approach is easy and flexible enough to be applied on different fields and it's a step forward into a standardize way of accessing large heterogeneous storage and computing facilities for complex scientific applications.